path: root/wiki/Chat_log/20170509-core-chatlog

Core-chat-meeting-2017-05-09

10:06 < geertu> Wecome to today's Core Group Chat
10:06 < geertu> s/Wecome/Welcome/
10:06 < geertu> Agenda:
10:06 < geertu> 1. Status updates
10:06 < geertu> Sort -R says I'm first
10:07 < geertu> A) What have I done since last time
10:07 < geertu>   - Published drivers/{clk,soc}/renesas/Kconfig rework
10:07 < geertu>   - Published R-Car Gen2 CPG/MSSR
10:07 < geertu>   - Published v2 of DTSI board sharing (Salvator-X and ULCB)
10:07 < geertu>   - Added more comments to periupport
10:07 < geertu> B) What I plan to do till next time
10:07 < geertu>   - Add more comments to periupport
10:07 < geertu>   - R-Car H3 ES2.0 DTS v2
10:07 < geertu> C) Problems I have currently
10:07 < geertu>   - What are the official names of the different versions of the R-Car
10:07 < geertu>     H3/M3 SiPs?
10:07 < geertu> D) Posted/Accepted bugfix patches
10:07 < geertu>   - None
10:07 < geertu> EOT
10:07 < geertu> Next is Jacopo
10:10 < geertu> Oh, he has no updates
10:10 < geertu> Next is Shimoda-san
10:10 < shimoda> yes
10:11 < shimoda> A) What have I done since last time
10:11 < shimoda> - Continue to investigate IPMMU workaround for Xen.
10:11 < shimoda>  - However, HW guy found that the workaround way is not good.
10:11 < shimoda>    So, he re-investigate the issue if we can have another way for workaround.
10:11 < shimoda> - No core task for linux kernel upstreaming.
10:11 < shimoda> B) What I plan to do till next time
10:11 < shimoda> - No Core task for next time.
10:11 < shimoda> C) Problems I have currently
10:11 < shimoda> - I have concern when IPMMU driver for Gen3 is merged into iommu subsystem repo.
10:11 < shimoda> D) Posted/Accepted bugfix patch
10:11 < shimoda> - Nothing.
10:11 < shimoda> EOT
10:12 < geertu> Thank you Shimoda-san. I assume we'll revise IPMMU when Magnus talks
10:12 < geertu> Next is Morimoto-san
10:13 < morimoto> I'm still under Renesas talk
10:13 < morimoto> sorry
10:13 < geertu> OK
10:13 < geertu> Let's postpone
10:13 [Users #periperi]
10:13 [ dammsan] [ jmondi  ] [ marex-cloud] [ neg      ] 
10:13 [ geertu ] [ kbingham] [ morimoto   ] [ pinchartl] 
10:13 [ horms  ] [ Marex   ] [ mturquette ] [ shimoda  ] 
10:13 -!- Irssi: #periperi: Total of 12 nicks [0 ops, 0 halfops, 0 voices, 12 normal]
10:13 < geertu> Ulrich is not here
10:13 < geertu> Next is (IP)M(M)agnUs
10:14 < dammsan> hehe
10:14 < jmondi> ups, sorry, I missed you have called my name Geert... No updates from me, yes
10:14 < dammsan> i've been doing this and that, including some MTU fixes for remote access,
10:15 < dammsan> getting Migo-R to boot mainline and also updating the IPMMU driver
10:15 < dammsan> i have no code to share about the IPMMU driver yet, but this is my main focus from now on
10:15 < neg> Why is Migo-R interessting?
10:15 < dammsan> i would like to ask for your help with PFC on Migo-R, but maybe we can deal with that later
10:16 < geertu> Sure
10:16 < dammsan> Migo-R has CEU devices for camera development
10:16 < neg> ahh
10:16 < dammsan> that's it from my side
10:17 < geertu> Thank you, Magnus
10:17 < geertu> IPMMU is getting more and more complicated.
10:17 < geertu> The trick with xlate() no longer works, as errors are now checked, thanks to a two year old patch from Laurent, which was picked up by someone else
10:18 < dammsan> thanks guys =)
10:18 < geertu> Next is Niklas
10:18 < neg> A) No core task since last time
10:18 < neg> B) Try to move '[PATCH 0/3] dmaengine: rcar-dmac: fix resource freeing synchronization' forward.
10:18 < neg> C) None
10:18 < neg> D) None
10:19 < neg> EOT
10:20 < geertu> Thank you, Ni9klas
10:20 < geertu> s/Ni9klas/Niklas/
10:21 < geertu> Next is Marek
10:21 < Marex> A) What have I done since last time
10:21 < Marex> - Submitted V2 of ROHM PMIC patches -> GPIO and Regulator bits in v4.12 -> MFD part waiting for Lee Jones (Mr. MFD), likely 4.13
10:21 < Marex> - Fixed GyroADC driver DT clock bindings
10:21 < Marex> - Prepared r8a7796 mainline U-Boot patches for submission
10:21 < Marex> B) What I plan to do till next time
10:21 < Marex> - Submit U-Boot patches adding r8a7796 support and Salvator-X M3 support for mainline inclusion (incl. USB, ethernet, SD/MMC support)
10:21 < Marex> C) Problems I have currently
10:21 < Marex> - NONE
10:21 < Marex> D) Posted/Accepted bugfix patch
10:21 < Marex> - Patch from Morimoto-san superseding my r8a7791 audio fix: https://patchwork.kernel.org/patch/9691497/
10:21 < Marex> EOT
10:21 < geertu> Thank you, Marek
10:22 < geertu> Next is Laurent
10:22 < geertu> lurking?
10:22 < pinchartl> yes :-)
10:22 < pinchartl> nothing to report for core
10:23 < geertu> Thank you, Laurent
10:23 < geertu> Last but not least is Simon
10:23 < horms> Todo Update
10:23 < horms>    NULL
10:23 < horms> A) What have I done since last time
10:23 < horms> B) What I plan to do till next time
10:23 < horms>    No Core task at this time
10:23 < horms> C) Problems I have currently
10:23 < horms> D) Posted/Accepted bugfix patch
10:23 < horms>    None
10:23 < horms> ---
10:24 < geertu> Thank you, Simon
10:26 < geertu> I guess Morimoto-san is still meeting, to negotiate his insurance. It would be really bad for Renesas business if something happened to him!
10:27 < geertu> Jacopo: Do you want something to say about RZ/A1 PFC?
10:27 < jmondi> eh
10:27 < jmondi> I would have a lot of things to say :p
10:28 < geertu> I think you're suffering from the "code arriving just too late for the merge window has 7 extra weeks for biekshedding
10:28 < geertu> s/biekshedding/bikeshedding syndrome"/
10:28 < jmondi> seriously, seems like Linus is now for "renesas,bi-directional" etc generic properties
10:29 < horms> its almost like some of these people don't have enough to do with their time
10:29 < jmondi> which is fine, if this means we can go forward on this... I will give some time to others to reply, and get back with a slightly different proposal
10:29 < jmondi> but -maybe- this will finally see the light of the day
10:30 < geertu> Let's hope so
10:30 < jmondi> indeed :)
10:30 < geertu> Does anyone else has something to discuss (I know Magnus and Migo-R PFC)
10:31 < geertu> ?
10:32 < neg> I have nothing
10:32 < horms> I was only wondering if anyone is waiting for me for anything - e.g. patches I have missed
10:33 < dammsan> horms: the recently PMIC driver and MAX chip from Marek and Jacopo want to get integrated
10:34 < dammsan> maybe you've got that sorted already
10:34 < dammsan> they will be in -rc1
10:34 < Marex> dammsan: I think only the GPIO and Regulator bits will be in rc1 ... Lee didn't pick the MFD part
10:34 < geertu> "arm64: dts: salvator-x: Add current sense amplifiers" is in Simon's tree
10:34 < Marex> (yet)
10:35 < dammsan> good then i'll be silent =)
10:35 < horms> likewise
10:37 < geertu> OK, I think we can move on to Magnus' PFC problem?
10:37 < dammsan> Some detail exists in recently sent email "Running upstream Linux on Migo-R"
10:37 < dammsan> The boot log includes the following error:
10:38 < dammsan> sh-pfc pfc-sh7722: pin 0 already registered
10:38 < dammsan> sh-pfc pfc-sh7722: error during pin registration
10:38 < dammsan> sh-pfc pfc-sh7722: could not register: -22
10:38 < dammsan> sh-pfc: probe of pfc-sh7722 failed with error -22i think it is relate to the SMSC ethernet error also listed
10:39 < dammsan> may i give this issue to the core group? =)
10:39 < geertu> There doesn't seem to be any message about Ethernet?
10:39 < geertu> I can give it a try
10:40 < geertu> Probably you have no idea which was the latest working version?
10:40 < dammsan> thanks
10:40 < dammsan> i was hoping that jacopo could help since he is the master of PFC
10:40 < dammsan> but it is really up to you guys how to handle it
10:40 < dammsan> i've tested a couple of different kernel versions
10:41 < dammsan> v2.6.34 v3.6.39 v3.19 v4.11
10:41 < dammsan> v2.6.39 is should be
10:42 < dammsan> v2.6.39 seems to work pretty well with the CEU and all according to the boot log
10:42 < dammsan> v4.11 requires disabling of the SMSC ethernet to proceed to the initramfs shell prompt
10:42 < jmondi> dammsan: not sh PFC, sorry :P
10:42 < geertu> Ah, you disabled Ethernet
10:43 < dammsan> yeah to move forward
10:43 < jmondi> dammsan: also please have a look at my proposal to use Peach/Genami for CEU :)
10:43 < dammsan> i think v3.19 may have output other PFC errors, don't recall exactly
10:43 < geertu> PFC worked in v2.6.39?
10:43 < dammsan> yes i believe so
10:43 < dammsan> mainline used to be fine about 5 years ago
10:44 < geertu> Probably it's a conflict between traditional fixed pin numbers and new DT dynamics
10:44 < dammsan> might be
10:44 < dammsan> jmondi: feel free to poke around with your own hardware, but don't ignore existing working platforms
10:45 < jmondi> dammsan: working you said? :P
10:45 < dammsan> yeah
10:45 < dammsan> its a software issue
10:45 < dammsan> on a known working platform
10:45 < dammsan> as opposed to unknown status on an unknown platform =)
10:46 < jmondi> eheh, we live on the edge!
10:46 < geertu> OK, I'll see if I can reproduce the problem
10:46 < dammsan> thanks, do you need anything from me?
10:46 < dammsan> you may want to use the same tool chain
10:47 < dammsan> (which i'm not 100% sure if it is correct to be honest)
10:47 < geertu> I have cross-compilers for about everything
10:47 < geertu> sh4-linux-gcc 4.6.3
10:47 < dammsan> great, let me know when sh7722 support gets merged upstream in gcc =)
10:47 < dammsan> and glibc =)
10:47 < dammsan> but anyway
10:48 < geertu> gcc-4.6.3-nolibc ;-)
10:48 < dammsan> i wish you the best of luck
10:48 < dammsan> i used to only stick to crosstool
10:48 < dammsan> then i was introduced to the hell also known as out-of-tree toolchain patches
10:49 < dammsan> i think sh7750 should have a good chance, but SH4ALDSP may not
10:49 < dammsan> work as-is out of the box
10:49 < dammsan> there is also half-assed FPU emulation in the kernel
10:49 < dammsan> let me know if you want the binary
10:51 < geertu> ok
10:51 < geertu> Anything else to discuss?
10:53 < jmondi> not from me!
10:53 < horms> geertu: do you have a cross compiler for ARCH=arc ?
10:53 < geertu> horms: No
10:54 < horms> np, it was a long shot
10:54 < geertu> I believe it's not available from kernel.org crosstool
10:54 < horms> it wasn't when I checked last week
10:55 < geertu> You can ask Arnd. If I send a patch for an issue with my old m68k gcc 4.1.2, he replies with a list of gcc versions that show the same warning ;-)
10:55 < horms> ok, thanks for the idea
10:55 < geertu> Thanks for joining, and have a nice continued day!